schrodinger.pipeline.stages.combine module

Core stages for Pipeline mechanism.

CombineStage

Stage for combining multiple structure PipeIO sets into one PipeIO set and labelling each ligand by the input set from which it came.

DataFusionMergeStage

Stage for merging results in a Data Fusion workflow.

Copyright Schrodinger, LLC. All rights reserved.

class schrodinger.pipeline.stages.combine.CombineStage(*args, **kwargs)

Bases: schrodinger.pipeline.stage.Stage

Stage for combining multiple Structures objects into one. Source object can be labeled by supplying LABEL keywords

The keywords specific to this stage are…

LABELFIELD An optional property added to each ouput structure

that holds the label for the input set whence it came.

LABELS List of labels for the input sets.

The stage takes up to ten input structure files sets and generates one output structure file set.

__init__(*args, **kwargs)

See class docstring.

operate()

Combine all the input files from all input sets into one set, optionally labelling each structure according to the set from which it originated. Raises a RuntimeError if there is a problem reading an input file or writing an output file.

JobDJOptions()

Returns a dictionary of options to pass to JobDJ: hosts, max_retries, default_max_retries, verbosity

__contains__(key)
__len__()
addExpectedInput(position, type, required=True)

A stage can accept one or more pipeio input objects. Use this method to specify the type of input object that is expected at each position.

position - an integer starting at 1. type - structures/grids/etc. required - whether this input always needs to be specified

addExpectedOutput(position, type, always=True)

A stage can return one or more pipeio objects. Use this method to specify the type of object that will be returned and whether or not it will always be produced by the stage.

position - an integer starting at 1. type - structures/grids/etc. always - whether this output is always produced

addOutputFile(filename)

Adds the specified file to the stage’s job control record. File must be specified as local (not absolute) path.

areSubjobsAndStageOnSameHost()

Return True if subjobs are running on the same host as the driver and the stages. This is always the case with JOB_SERVER but not with classic Job Control.

checkFile(file, error='File does not exist:')

Raise exception if specified file does not exist. The message that is printed can be specified.

checkFiles(files, error='File does not exist')

Raise expetion if any file does not exist.

checkInputs()

OVERWRITE: Return False if something is wrong with the input files or the parameter, otherwise return True.

checkParameters()

OVERWRITE: Make sure that all parameters are valid.

checkProducts()

Raises RuntimeError if any of the required products are not installed or the version installed is less that minimum required version. It is possible to override this method. See ligprep.py for example.

clear() None.  Remove all items from D.
copy()
debug(text)

Print a debug line to the log file

dump()

This method dumps all the variables of the Stage to a restart file. Run it every time an important step is performed.

error(text)

Print an error line to the log file

exit(text='')

Print an error line to the log file and exit with code 1

classmethod fromkeys(iterable, value=None)
genFileName(extension=None, filenum=None, start=None, end=None)

Generate a file name to be used by the stage. Returns string:

"<full-stagename>-<filenum><extension>"
"<full-stagename>-<start>_<end><extension>"
"<full-stagename><extension>"
"<full-stagename>", etc.

Depending on given options.

genOutputFileName(position, extension='', filenum=None, start=None, end=None)

Generate a file name to be used by the stage when writing files for the output position <position>. Returns strings:

"<full-varname>-<filenum><extension>"
"<full-varname>-<start>_<end><extension>"
"<full-varname><extension>"
"<full-varname>", etc.

Depending on given options.

get(k[, d]) D[k] if k in D, else d.  d defaults to None.
getAdjustedNJobs(total_mol, min_job_size, max_job_size)

Returns the desired number of subjobs, and adjusts it for the the specified min & max job sizes if the user specified ADJUST option. If the number of desired jobs was specified by the user, the number of available cpus is used or 10, whichever is smaller. Specify the number of input ligands and the smallest and largest desired job sizes (Generally job lengths of 1 minute & 24 hours).

getCleanupRequested()

Stages should clean up after themselves if this returns True

getHostList()

Returns a list of hosts to run the subjobs on. localhost:1 may be in the list as well. Ideally, pass the output to JobDJ. Format: [ (host1,ncpus), (host2,ncpus) ] Pass this value to JobDJ.

getHostStr()

Just like getHostList() but instead of returning a list, returns a host string to be passed to the -HOST argument.

getInput(position)

Use in operate() to get the input object for specified position. Returns None if invalid position is specified.

getInputNames()

Return a dictionary of variable name of the inputs at each position. Key:position, value:name

getJobDJ(**kwargs)

Returns a pre-set JobDJ instance for the stage to use. It already has it’s hosts, max_retries, max_failures, default_max_retries, and verbosity set.

getMaxRetries()

Return the number of max restarts to use. If -max_retries is specified, returns that value; otherwise if SCHRODINGER_MAX_RETRIES is defined, returns that value; otherwise returns default of 2. Pass this value to JobDJ.

getNCpus()

Returns the total number of processors specified in the host string. For queued hosts with no CPU# specification, 10 is added.

getNJobs()

Returns the requested target number of subjobs, and whether or not to adjust that number if it is unreasonable.

If -NJOBS was not specified, the # of CPUs or 10 is returned (whichever is smaller).

Used by Glide DockingStage and _adjustNJobs()

getName()

Return stagename (jobname of the stage)

getOutput(position)

Returns the output IO object of the stage at specified position. Use this method after running the stage to get its output objects

getOutputName(position)

Return the output name for specified position

getOutputNames()

Return a list of output names for each position (dict)

getRuntimePath(filename)

Return the runtime-path of a file that user specified Prints an error and exits if file does not exist.

getStageDirectory()

Return the directory in which the stage is running

getVerbosity()

Return verbosity of thos stage (for JobDJ)

hasCompleted()

Returns True if this stage’s operate() exited successfully.

hasStarted()

Returns True if this stage has started.

info(text)

Print an info line to the log file

initNonPersistent(pipeline)

Gets called after the Stage joins pipeline. Meant to be used to initialize non-persistent context.

items() a set-like object providing a view on D’s items
iterInputs()

Iterate through input objects: (position, obj)

keys() a set-like object providing a view on D’s keys
lognoret(*args)

Prints specified objects to the stage log file. No EOF return

mainProduct()

If a stage has a main product associated with it, the stage should overwrite this method with a method that returns the product string. For example, the LigPrepStage.mainProduct() will return “ligprep” Used by Pipeline.

outputRequested(position)

Returns True if the user requested optional output at <position>

pop(k[, d]) v, remove specified key and return the corresponding value.

If key is not found, d is returned if given, otherwise KeyError is raised.

popitem() (k, v), remove and return some (key, value) pair

as a 2-tuple; but raise KeyError if D is empty.

reportParameters(fh=None)

Print the value of each keyword for this stage to the stream specified as <fh>. Used by Pipeline

requiredProduct(product)

Specify a product that is required for this stage to run; optionally minimum version.

Example: product=”mmshare”

requiredProductRuntime(product)

Similar to requiredProduct() but can be used to specify required products at runtime. For example, ConvertStage doesn’t know what products are required for conversion until runtime. Raises RuntimeError if product is not installed.

run(idle_function=None, restart_file=None, verbosity=None, logfh=None)

Run the stage.

idle_function - function to call when idle

restart_file - file to periodically dump this instance to

verbosity - there are three verbosity levels: “quiet”, “normal”, and “verbose”

“quiet” - only warnings and errors are printed “normal” - stage progress is printed - default “verbose” - additional debugging info is printed

logfh - where to send the loggin output

runJobDJ(jobdj)
setInput(position, name=None, obj=None)

Specify an input to use for this stage. position - input specified is for this position name - Variable name of this IO object obj - the IO object

This method is called by Pipeline.

setJobDJOptions(jobdj)

Use this method to adjust the specified queue.JobDJ instance to the VSW settings.

setJobOptions(subjob_hosts=None, njobs=None, adjust=None, force=None, max_retries=None, cleanup=None)

Tell this stage how to run the subjobs

Parameters
  • subjob_hosts – list of hosts to run subjobs on

  • njobs – number of subjobs to generate. None means determine automatically.

  • adjust – whether to adjust njobs such that job size is within limits

  • force – whether to continue with job if subjobs fail

  • max_retries – number of times to attempt to restart a subjob If not specified, use SCHRODINGER_MAX_RETRIES or 2.

  • cleanup – whether to delete intermediate files

setMainProduct(product)

Specify which product this stage is part of. Will determine which host the subjobs are run on.

setOutput(position, obj)

Use this method at the end of operate() to set the output.

setOutputName(position, varname)

Is called by Pipeline when starting the stage. Tell the stage what name to save each output under.

setdefault(k[, d]) D.get(k,d), also set D[k]=d if k not in D
update([E, ]**F) None.  Update D from mapping/iterable E and F.

If E present and has a .keys() method, does: for k in E: D[k] = E[k] If E present and lacks .keys() method, does: for (k, v) in E: D[k] = v In either case, this is followed by: for k, v in F.items(): D[k] = v

updateJobdj(jobdj)

Gets called periodically in order to update JobDJ’s hosts. Will ask Pipeline for CPUS when needed, and will tell Pipeline when they are no longer needed.

validateValues(preserve_errors=False)

Validates the stored keywords. This is done by converting <self> to a ConfigObj instance, and calling validate() on it. The validated keywords are then updated back to <self>. This is done as part of Ev:87429

values() an object providing a view on D’s values
warning(text)

Print a warning line to the log file

class schrodinger.pipeline.stages.combine.DataFusionMergeStage(*args, **kwargs)

Bases: schrodinger.pipeline.stage.Stage

This stage takes in results from a Glide docking job, Phase Shape job, and a Canvas 2D fingerprint job and combines them based on the Z-score for each method.

This stage is used by Data Fusion workflow (data_fusion_backend.py).

Z-score = the number of standard deviations above or below the mean of the

distribution of scores for that method.

Final score = For each compound, the sum of z-scores for the 3 methods.

Compounds will be sorted by the final score in the output.

STAGE Merge Results

Input 1: Docking PV file Input 2: Phase Shape output file Input 3: Canvas 2D Fingerprints output file Output: Merged Maestro file of compounds sorted by the final score.

Properties included in the output: 1. Glide score 2. Phase Shape score 3. Similarity score 4. Consensus/final score (average of z-scores)

WARNING: This stage assumes that only one ligand with the same UNIQUEFIELD exists in each input set.

__init__(*args, **kwargs)

Creates the stage instance, and passes the <args> and <kwargs> to the stage.Stage’s constructor.

operate()

The only overridden & required method in this class. Called by the Pipeline to run this stage’s main code.

calcZScores(scores_by_compound, more_negative_is_better=False)

Takes in a dictionary where keys are compound IDs, and values are scores, and returns a dict of z-scores (keys are compound IDs also). Z-score is calculated by:

z-score = (score-average) / std-deviations

If the number of compounds is 1, then the z-score will be <None>.

Parameters
  • scores_by_compound (dict) – Dictionary of scores (e.g. Glide score) keyed by compound ID.

  • more_negative_is_better (bool) – If set to True, more negative scores are considered to be better, and will result in higher Z-scores.

Returns

Z-Scores, in a dictionary keyed by compound ID.

Return type

dict

calcConsensusScore(scores, top_n)

Given a list of scores, select the best <top_n> of them, and calculate their average.

JobDJOptions()

Returns a dictionary of options to pass to JobDJ: hosts, max_retries, default_max_retries, verbosity

__contains__(key)
__len__()
addExpectedInput(position, type, required=True)

A stage can accept one or more pipeio input objects. Use this method to specify the type of input object that is expected at each position.

position - an integer starting at 1. type - structures/grids/etc. required - whether this input always needs to be specified

addExpectedOutput(position, type, always=True)

A stage can return one or more pipeio objects. Use this method to specify the type of object that will be returned and whether or not it will always be produced by the stage.

position - an integer starting at 1. type - structures/grids/etc. always - whether this output is always produced

addOutputFile(filename)

Adds the specified file to the stage’s job control record. File must be specified as local (not absolute) path.

areSubjobsAndStageOnSameHost()

Return True if subjobs are running on the same host as the driver and the stages. This is always the case with JOB_SERVER but not with classic Job Control.

checkFile(file, error='File does not exist:')

Raise exception if specified file does not exist. The message that is printed can be specified.

checkFiles(files, error='File does not exist')

Raise expetion if any file does not exist.

checkInputs()

OVERWRITE: Return False if something is wrong with the input files or the parameter, otherwise return True.

checkParameters()

OVERWRITE: Make sure that all parameters are valid.

checkProducts()

Raises RuntimeError if any of the required products are not installed or the version installed is less that minimum required version. It is possible to override this method. See ligprep.py for example.

clear() None.  Remove all items from D.
copy()
debug(text)

Print a debug line to the log file

dump()

This method dumps all the variables of the Stage to a restart file. Run it every time an important step is performed.

error(text)

Print an error line to the log file

exit(text='')

Print an error line to the log file and exit with code 1

classmethod fromkeys(iterable, value=None)
genFileName(extension=None, filenum=None, start=None, end=None)

Generate a file name to be used by the stage. Returns string:

"<full-stagename>-<filenum><extension>"
"<full-stagename>-<start>_<end><extension>"
"<full-stagename><extension>"
"<full-stagename>", etc.

Depending on given options.

genOutputFileName(position, extension='', filenum=None, start=None, end=None)

Generate a file name to be used by the stage when writing files for the output position <position>. Returns strings:

"<full-varname>-<filenum><extension>"
"<full-varname>-<start>_<end><extension>"
"<full-varname><extension>"
"<full-varname>", etc.

Depending on given options.

get(k[, d]) D[k] if k in D, else d.  d defaults to None.
getAdjustedNJobs(total_mol, min_job_size, max_job_size)

Returns the desired number of subjobs, and adjusts it for the the specified min & max job sizes if the user specified ADJUST option. If the number of desired jobs was specified by the user, the number of available cpus is used or 10, whichever is smaller. Specify the number of input ligands and the smallest and largest desired job sizes (Generally job lengths of 1 minute & 24 hours).

getCleanupRequested()

Stages should clean up after themselves if this returns True

getHostList()

Returns a list of hosts to run the subjobs on. localhost:1 may be in the list as well. Ideally, pass the output to JobDJ. Format: [ (host1,ncpus), (host2,ncpus) ] Pass this value to JobDJ.

getHostStr()

Just like getHostList() but instead of returning a list, returns a host string to be passed to the -HOST argument.

getInput(position)

Use in operate() to get the input object for specified position. Returns None if invalid position is specified.

getInputNames()

Return a dictionary of variable name of the inputs at each position. Key:position, value:name

getJobDJ(**kwargs)

Returns a pre-set JobDJ instance for the stage to use. It already has it’s hosts, max_retries, max_failures, default_max_retries, and verbosity set.

getMaxRetries()

Return the number of max restarts to use. If -max_retries is specified, returns that value; otherwise if SCHRODINGER_MAX_RETRIES is defined, returns that value; otherwise returns default of 2. Pass this value to JobDJ.

getNCpus()

Returns the total number of processors specified in the host string. For queued hosts with no CPU# specification, 10 is added.

getNJobs()

Returns the requested target number of subjobs, and whether or not to adjust that number if it is unreasonable.

If -NJOBS was not specified, the # of CPUs or 10 is returned (whichever is smaller).

Used by Glide DockingStage and _adjustNJobs()

getName()

Return stagename (jobname of the stage)

getOutput(position)

Returns the output IO object of the stage at specified position. Use this method after running the stage to get its output objects

getOutputName(position)

Return the output name for specified position

getOutputNames()

Return a list of output names for each position (dict)

getRuntimePath(filename)

Return the runtime-path of a file that user specified Prints an error and exits if file does not exist.

getStageDirectory()

Return the directory in which the stage is running

getVerbosity()

Return verbosity of thos stage (for JobDJ)

hasCompleted()

Returns True if this stage’s operate() exited successfully.

hasStarted()

Returns True if this stage has started.

info(text)

Print an info line to the log file

initNonPersistent(pipeline)

Gets called after the Stage joins pipeline. Meant to be used to initialize non-persistent context.

items() a set-like object providing a view on D’s items
iterInputs()

Iterate through input objects: (position, obj)

keys() a set-like object providing a view on D’s keys
lognoret(*args)

Prints specified objects to the stage log file. No EOF return

mainProduct()

If a stage has a main product associated with it, the stage should overwrite this method with a method that returns the product string. For example, the LigPrepStage.mainProduct() will return “ligprep” Used by Pipeline.

outputRequested(position)

Returns True if the user requested optional output at <position>

pop(k[, d]) v, remove specified key and return the corresponding value.

If key is not found, d is returned if given, otherwise KeyError is raised.

popitem() (k, v), remove and return some (key, value) pair

as a 2-tuple; but raise KeyError if D is empty.

reportParameters(fh=None)

Print the value of each keyword for this stage to the stream specified as <fh>. Used by Pipeline

requiredProduct(product)

Specify a product that is required for this stage to run; optionally minimum version.

Example: product=”mmshare”

requiredProductRuntime(product)

Similar to requiredProduct() but can be used to specify required products at runtime. For example, ConvertStage doesn’t know what products are required for conversion until runtime. Raises RuntimeError if product is not installed.

run(idle_function=None, restart_file=None, verbosity=None, logfh=None)

Run the stage.

idle_function - function to call when idle

restart_file - file to periodically dump this instance to

verbosity - there are three verbosity levels: “quiet”, “normal”, and “verbose”

“quiet” - only warnings and errors are printed “normal” - stage progress is printed - default “verbose” - additional debugging info is printed

logfh - where to send the loggin output

runJobDJ(jobdj)
setInput(position, name=None, obj=None)

Specify an input to use for this stage. position - input specified is for this position name - Variable name of this IO object obj - the IO object

This method is called by Pipeline.

setJobDJOptions(jobdj)

Use this method to adjust the specified queue.JobDJ instance to the VSW settings.

setJobOptions(subjob_hosts=None, njobs=None, adjust=None, force=None, max_retries=None, cleanup=None)

Tell this stage how to run the subjobs

Parameters
  • subjob_hosts – list of hosts to run subjobs on

  • njobs – number of subjobs to generate. None means determine automatically.

  • adjust – whether to adjust njobs such that job size is within limits

  • force – whether to continue with job if subjobs fail

  • max_retries – number of times to attempt to restart a subjob If not specified, use SCHRODINGER_MAX_RETRIES or 2.

  • cleanup – whether to delete intermediate files

setMainProduct(product)

Specify which product this stage is part of. Will determine which host the subjobs are run on.

setOutput(position, obj)

Use this method at the end of operate() to set the output.

setOutputName(position, varname)

Is called by Pipeline when starting the stage. Tell the stage what name to save each output under.

setdefault(k[, d]) D.get(k,d), also set D[k]=d if k not in D
update([E, ]**F) None.  Update D from mapping/iterable E and F.

If E present and has a .keys() method, does: for k in E: D[k] = E[k] If E present and lacks .keys() method, does: for (k, v) in E: D[k] = v In either case, this is followed by: for k, v in F.items(): D[k] = v

updateJobdj(jobdj)

Gets called periodically in order to update JobDJ’s hosts. Will ask Pipeline for CPUS when needed, and will tell Pipeline when they are no longer needed.

validateValues(preserve_errors=False)

Validates the stored keywords. This is done by converting <self> to a ConfigObj instance, and calling validate() on it. The validated keywords are then updated back to <self>. This is done as part of Ev:87429

values() an object providing a view on D’s values
warning(text)

Print a warning line to the log file